Gourav Siddhad, Yogesh Kumar Meena
Hybrid BCIs combine two kinds of signal: motor imagery, where someone imagines moving, and steady-state visual evoked potentials, the rhythmic response to a flickering stimulus. Together they carry more information than either alone, and together they cost more to decode than embedded hardware can afford.
That gap decides what gets built. A decoder needing a workstation confines the interface to a lab bench, and the applications people actually want are wearable and battery-powered.
SwitchBraidNet targets the constraint directly, and the quantisation-aware part of the name matters. A model shrunk after training often loses accuracy in ways nobody anticipated; one trained knowing it will run in reduced precision can adapt to that during learning. The dual-path temporal braid reflects the same pragmatism, since the two signal types live at different timescales and forcing them through one pathway wastes capacity reconciling them.
Hybrid brain-computer interfaces (BCIs) that integrate motor imagery (MI) and steady-state visual evoked potentials (SSVEP) provide high-dimensional neural decoding but typically exceed the computational limits of embedded hardware. To address this, we propose SwitchBraidNet, a compact EEG classification architecture designed for low-power deployment. The model employs a dual-path temporal braid to extract multiscale oscillatory features, an adaptive squeeze-and-excitation…
EEG-Based Imagined Speech Decoding Using a Hybrid CNN-SNN Architecture
arXiv (neural decoding) · July 3, 2026Rethinking Brain Decoding with CLIP: The Role of Adversarial Robustness
arXiv (BCI) · July 3, 2026Stacked LoRA for Subject-Adaptive EEG Foundation Models in Motor Imagery Decoding
arXiv (BCI) · July 2, 2026Antenna System for Simultaneous Wireless Power and Information Transfer to Brain Implants